Nenad Gubeljak
Associate Professor
University of Maribor
Faculty of Mechanical Engineering
Mustafa Koçak
European Fitness-for-Service Network –
FITNET coordinator
GKSS Research Centre
Institute for Materials Research, Geesthacht
Michell Huther
Bureau Veritas, Paris
Tomaz Valh
Walch d.o.o, Maribor
FITNET Fitness-for-service Fracture
Module SOFTWARE
The paper gives concepts of FITNET software for Fracture module. In
order to perform structure integrity assessment for component with real or
assumed crack, it is necessary to vary material properties and crack
geometry. Such approach can provide statistically reliable results for
maximum loading capacity vs. crack size. Software can be useful and
simple tool which helps to avoid calculation of polynom’s parameters in
equations for limit loads and stress intensity factors. The new concept of
software is useful for education and introduction to FFS procedure in
company and departments with base engineering knowledge.
Keywords: fitness-for-purpose, structural integrity, FITNET.
1. INTRODUCTION
Today, decent software for structure integrity is
available [1,2]. Usually, the application of such software
and interpretation of results require good knowledge
about fracture mechanics, mechanics and materials.
Within the framework of FITNET project, a Fracture
education version is developed. The aim of the software
is not to replace the mentioned softwares. The main goal
of the software is to spread application to user with
different knowledge level and skills. It is acheived with
friendly operate enviroment, simple explainations and
blocked unnecessary operations and steps. With this
approach, the shortcut between input data and results is
established. This helps the beginners and students to
start to apply Fitness-for-Service concept for structure
in design, manufacturing and service. However, usage
of software helps to establish a new philosophy
regarding to structure with flaws or defects. Only few
case studies are included in the software.
Software is established with the module structure
which incorporates each component as a module with
known limit load solutions and stress intensity factor
solution. In the software SINTAP concept [3] is applied
for structure integrity analysis. Software checks validity
conditions for solutions and provides possibility to
change stress intensity factor with new solution or add a
new component. The material module includes routine
for input mechanical properties and fracture toughness.
The output routine enables to print and safe results as
report or plots and text file appropriate for further
analysis.
2. FITNET FFS FRACTURE MODULE
The FITNET procedure represents with its sequential
approach an ideal basis for computer manipulation. The
procedure might be seen as a collection of
recommendations meant for engineering purposes in
industry mainly for engineers with practical experiences
Received: March 2008, Accepted: April 2008
Correspondence to: Dr Nenad Gubeljak
University of Maribor, Faculty of Mech. Engineering,
Smetanova 17, SI-2000 Maribor, Slovenia
E-mail: [email protected]
© Faculty of Mechanical Engineering, Belgrade. All rights reserved
but limited knowledge of fracture mechanics. This
application considers only the standard options of
FITNET procedure; these are Default option, Basic
option, Mis-match option and Advanced option. When
increasing the option, the demand for the entry
parameters of the material increases. There's also
another issue, on which this application depends in
addition to the “standard” FITNET procedure. This is
the one of available methods (e.g. ETM [4], ETM-MM
[5], BS7910 [6], R6 [7]) for assessing the significance
of crack-like defects in engineering practice that offers
the results of the stress intensity factor and limit loading
for different configurations. The application already
contains a few examples.
The main requirement when planning the application
was the universality and openness of the tool for
manifold of elementary configurations in addition to the
already included ones. One should also be able to export
the results, data and graphs in standard formats into
other standard applications.
Though the FITNET procedure is standardised, each
solution holds for an individual configuration, on the
other hand. Its result in the simplest form is only an
information whether the construction is safe or there is a
potential danger of a failure (i.e. fracture). This is basic
information only. Consider that a crack can be safe for
the prescribed load but unsafe in case of overloading.
The application enables repeating the calculations with
different properties of the crack or different loadings for
that purpose. It remains up to the user to decide which is
the critical parameter.
The application is built up from two program
languages that coincide with the philosophy of
universality of the tool with respect to the user and
configurations. While MS Visual Basic serves mainly as
a graphical interface, C++ is used for loading additional
geometrical and loading configurations. The latter
enables the new configurations to be added on-line
without the need of annoying compilation. The
configurations are usually based on problem solving;
the FITNET procedure option depends on the
availability of the parameters of the material. The
boundary values (conditions) are selected by the user,
while the validity conditions depend on the stress
intensity factor and limit loading.
FME Transactions (2008) 36, 39-44
39
The calculation is performed in two phases. The first
one is the preparation of the stress intensity factor and
limit loading. The second phase is a standard FITNET
procedure carried out in FAD or CDF diagram.
3. GEOMETRICAL AND LOADING CONFIGURATION
Selection of the geometrical and loading configuration
is the basis for the calculation. At the left side of the
dialog all registered configurations with their number
and description are listed. Most of the dialogs are set to
the default values, therefore the history of previous
calculations is erased when a new configuration is
selected.
8.1 Entering geometry and loading parameters
4. ENTERING THE MATERIAL PARAMETERS
This dialog contains different fields to enable entering
the material properties. There are several options which
depend on the option of the FITNET procedure. In
addition to other material parameters, the information
about the shape of stress-deformation curve has to be
given (whether material exhibits continuous yielding or
Lüders plateau).
Parameters of the material with respect to the
FITNET procedure option are given hereafter:
Description
Unit
Young's Modulus of
Elasticity, E
Option
D
B MM A
GPa
9
9
9
9
Poisson's ratio, ν
–
9
9
9
9
Charpy impact energy, Cν
J
9
Working temperature, T
ºC
9
Temperature at 28J of Charpy
impact energy, T28J
ºC
9
Probability of failure, Pf
%
9
MPa
9
9
9
9
MPa
9
9
9
N/mm1.5,
N/mm,
mm
9
9
9
or
Yield strength at 0.2 % plastic
strain, Rp0.2
Tensile strength, Rm
Crack fracture resistance in
terms of K, J or δ (Kmat, Jmat,
δmat)
Engineering stressdeformation curve from single
axis tensile test, R-e
File
9
D – Default option; MM – Mis-match option; B – Basic option
A – Advanced option.
The parameters of the material are valid at the
operating temperature. When entering the material data
the dialog at the FITNET advanced option behaves in a
slightly different way. All data fields except the crack
fracture resistance as a material property are locked.
This is due to the fact that all material parameters are
entered into a special dialog together with engineering
(R-e) stress-deformation curve.
The engineering stress-deformation curve is needed
at FITNET Advanced option. The presence of boundary
conditions depends on an individual geometrical and
40 ▪ VOL. 36, No 1, 2008
loading configuration. These conditions are mainly user
selectable and define equations for stress intensity factor
and limit load calculation.
There are maximum three groups of different
boundary conditions:
• Stress/Strain condition (plane stress or plane
strain);
• Geometrical point;
• Geometry or Shape.
Each of the boundary condition groups can have at
the most three different options for selection. This
dialog is not shown if a particular configuration doesn’t
have boundary conditions to select.
The prepared FITNET application supports external
loads in terms of forces, momentums or arbitrary stress
distributions. The residual, internal or secondary
stresses are optional in calculation. The dialog can be
used in two different ways with respect to the method of
describing the load.
8.2 Simple loading case
The first parameter or the first two parameters are
considered as the key parameters of geometry or
loading. Considering the geometry, these are the crack
parameters. The crack size can be described only by
length or by length (c) and depth (a). These are the key
loading parameters. A simple load is usually described
by one parameter (e.g. force or momentum); combined
loads are often given by two parameters (e.g. tensile and
bending stress). There’s an optional selection right from
the first two geometrical and the first two loading
parameters, which represents a decision for primary and
secondary parameter of the crack geometry and loading.
The optional selection isn’t always available. When the
crack growth curve or loading curve is drawn, only the
selected primary parameter is increasing while the
secondary remains constant.
8.3 Distributed stress case
In this case the arbitrary stress distribution (ASD)
profile through the cross section of the component is the
loading input information. There are two more
command buttons available in addition to the common
ones. The numerical values are simply entered into the
corresponding data fields. Proper units have to be
considered. The verification of the values is performed
upon confirmation. With arbitrary stress distribution
profile the elementary loads, e.g. force, momentum or
more often membrane σm and bending σb stress and
polynomial coefficients of the stress distribution profile
(S0, S1, S2, ...) are present. These are not entry but
intermediate calculated data. However, these values can
also be inserted manually. There’s also a more reliable,
applicable and more often used choice. This dialog
enables the entering of the ASD profile through the
uncracked cross-section of the component. The stress
distribution profile is always referring to a structural
component without any crack. The stress distribution
FME Transactions
profile is given in a form of individual points – usually
along the thickness of the component cross-section. The
stress distribution profile is always parallel to one of the
characteristic dimensions of the component, while the
direction of the stress is orthogonal to the crack surface.
The position of the stress point is to be given always as
a ratio with respect to the characteristic dimension of
the cross-section, e.g. the position x with respect to the
thickness T (x/T). The points of the profile are not
completely arbitrary: a specific point has to be placed
within the limits of the chosen characteristic dimension,
e.g. thickness. The stress distribution profile first point’s
position is usually 0, the last point’s position is 1.
5. VERIFICATION OF GEOMETRICAL AND STRESS
VALIDITY CONDITIONS
The validity conditions discussed here are related to the
specific geometrical and loading configuration and are to
be used only within this configuration. The conditions
from stress intensity factor and limit load solution
originate the corresponding equations. The solutions
differ from one to another geometrical and loading
configuration. The conditions also differ due to different
solution. The details and explanation of individual
conditions are available in the original stress intensity
factor and limit load solutions compendium. Each
condition is described by mathematical expression or by
description. The result expresses the fulfilment of the
condition. The individual validity conditions are statically
verified for the operating point only, i.e. for the nominal
geometry and load. They aren’t verified through the
calculation with the crack growth or with increasing load
when loading curve or crack growth curve is drawn.
Therefore, though the nominal conditions are fulfilled,
they might not be fulfilled far from the operating point.
For that reason, the calculation has to be repeated for the
critical crack length and critical load as a new operating
point. This is the main advantage that allows the
computer software calculation tool.
6. ENTERING THE CALCULATION PARAMETERS
This dialog enables decision about the assessment with
respect to the critical load or critical crack (Critical
value assessment). The calculation is repeated with only
one parameter being changed, while the other remains
constant. Thus the loading curve or the crack growth
curve is drawn. This method enables us to determine not
only the operating or nominal point, but also the critical
or failure point. There are two parameters that define the
increase of selected value, the step for increasing the
load or crack length and the maximum load or crack
length. The lower limit of the increasing parameter is
selected as a small positive number. The step for
increasing the normalized load Lr has to be given, too.
The lower limit of the increasing normalized load is
always zero, while the plastic collapse limit defines the
upper limit. With this dialog the final act of the
parameter entry is presented. It usually erases the
history of previous calculations, if they exist. The
history of calculations can be preserved for future
comparisons by means of the control, but this can only
FME Transactions
be done within one geometrical and loading
configuration. The above mentioned control is disabled
whether there is no history or whether this is the first
calculation. The history is also erased with a choice of a
new geometrical and load configuration or if one
reselects the same configuration. The plastic collapse
limit of the normalised load is displayed in this dialog.
It usually cannot obtain the value that’s less than one. If
under certain conditions it becomes less than one, the
red colour indicates a probable fault. A selection of the
calculation with respect to the critical load or crack is in
case of an arbitrary stress distribution profile limited to
the crack growth and therefore the increasing parameter
cannot be freely chosen in the dialog.
7. GRAPHIC INTERPRETATION OF RESULTS
The basic purpose of the FITNET procedure and the
graphical representation of the results is the assesment
of the structural integrity of the component. The
location of two points is determined with the use of
various calculation methods. The first point is called the
nominal or operating point, while the second is known
as a critical or failure point. Both are presented in both
graphs (FAD and CDF) and give identical results. The
position of the nominal point with respect to the failure
curve is of our interest. This indicates the failure safety
of the structural component. The failure point expresses
the critical state, having two parameters: critical crack
length and the failure load.
After the last dialog is confirmed, the FITNET
calculation is performed in two stages. The first stage is
performed in Python environment. Firstly, the selected
parameter is increased, then the calculation of stress
intensity factor, limit load, reference stress and
normalised load is performed. The data are transferred
into the main application before the second stage
follows. The second stage is the standard FITNET
procedure with prescribed equations. Their selection
depends on the option of the procedure and on the type
of material yielding (Lüders plateau). The application
pop-ups a short message box with a results summary.
The message box also contains the coordinates of
nominal and critical point with respect to both
interpretation types and the critical length of the crack
or failure load, obtained from the graphs.
8. APPLICATION OF SOFTWARE
To illustrate the use and the possibilities offered by the
software, we shall present a case built up from an
industrial case. The 3 options presented in FITNET
documents are applied with the same service conditions.
8.1 Case study
A project for a Canadian Arctic harbour requires the
design of a steel box beam structure which dimensions
are: width = 2.0 m, height = 1.5 m, thickness 25 mm.
The designer selects an HS steel with the following
characteristics:
• Yield Strength 410 MPa;
• Ultimate Strength 630 MPa.
VOL. 36, No 1, 2008 ▪ 41
The stress-strain curve is known and given by the
following measurement data (σ in MPa):
ε
σ
0 0.002 0.01 0.02 0.04 0.07 0.08 0.09 0.1 0.12
0 410 460 510 560 590 600 610 620 630
The applied loads are an axial force and 2 local
loads (Fig. 1) leading at the weld toe to the following
stresses:
global plus local tension: 310 MPa
local plate bending: 20 MPa
(tension at the outer tube surface)
The operational environment conditions in winter
correspond to a minimum air temperature of – 50 °C.
F
weld
F
The beam integrity will be continuously checked by
internal pressure monitoring.
The raised question is what brittleness requirements
have to be introduced in the steel specification to
prevent the risk of brittle fracture before leakage or
ductile failure in service.
The 3 options of risk of brittle fracture presented in
FITNET will be applied using the developed above
presented demo software. The presented example
corresponds to the “Surface Cracks in Plates” (ASD)
case.
8.2 Default Option
The Default option assessment requires only to know
the steel yield strength and the service temperature and
allows only to calculate C or T28J temperature for a
given crack size.
For the application we decided to determine the
necessary T28J value.
The T28J is found equal to – 88 °C without brittle
fracture and with a critical crack length of 19.1 mm.
8.3 Basic Option assessment
Figure 1. Steel box beam shape and loadings
The return experience of similar structures leads to
assume that surface cracks can develop without being
detected at the weld toe and that their shape factor is a/c
= 0.2.
The Basic option assessment requires to know the steel
yield and ultimate strength, the service temperature and
allows to determine Kmat or J or the CTOD.
For the application we decided to determine the
necessary CTOD value.
To start a crack size is fixed, equal to a = 5 mm and
a CTOD value is given equal to 0.1.
Figure 2. FAD curve for Default option limit state
42 ▪ VOL. 36, No 1, 2008
FME Transactions
With the same stress distribution data than
previously, a first calculation provides a critical crack
length of 11.8 mm.
This value does not correspond to a fail safe
condition, the crack length is smaller than the plate
thickness, so brittle fracture will occur before leakage.
Using the “Repeat calculation” function, the CTOD
value will be changed until obtaining by iteration a
failure point on the limit state curve. The FAD is given
in the Figure 3.
The CTOD is found equal to 0.75 without brittle
fracture and with a critical crack length of 16.5 mm.
8.4 Advanced Option assessment
The Advanced option, as the Basic option assessment,
allows to determine Kmat or J or the CTOD, requires the
service temperature but also to know the steel stressstrain curve.
For the application we decided to determine the
necessary CTOD value.
The stress-strain data are entered, but due to their
small number, the “Approximate engineering R-e
curve” function is applied.
To start a crack size is fixed, equal to a = 5 mm and
a CTOD value is given equal to 0.1.
With the same stress distribution data than
previously, a first calculation provides a critical crack
length of 14.1 mm.
This value does not correspond to a fail safe
condition, the crack length is smaller than the plate
thickness, so brittle fracture will occur before leakage.
Using the “Repeat calculation” function, the CTOD
value will be changed until obtaining by iteration a
failure point on the limit state curve. The FAD is given
in the Figure 4.
The CTOD is found equal to 0.25 without brittle
fracture and with a critical crack length of 16.5 mm.
8.5 Results summary
The application of the 3 FITNET´s option provides the
following requirements for the steel brittleness
specification:
Default Option
minimum Kv = 28 J at – 88 °C
Basic Option
minimum CTOD = 0.75 mm at – 50 °C
Advanced Option
minimum CTOD = 0.25 mm at – 50 °C
9. CONCLUSIONS
The application of the software of the Fracture module
of the FITNET FSS Procedure to real case gives simple
but important information about materials properties
requirements. The software is possible to use in design
stage (choice of right material) and in-service,
(evaluation of risks from detected cracks). Software is
established with module structure which incorporates
each component as module with known limit load
solutions and stress intensity factor solution. In the
software the FITNET concept [3] is applied for structure
integrity analysis. Software checks valid conditions for
Figure 3. FAD curve for Basic option limit state
FME Transactions
VOL. 36, No 1, 2008 ▪ 43
Figure 4. FAD curve for Advance option limit state
solutions and provides possibility to change stress
intensity factor with a new solution or adds new
components. Material module includes routine for input
mechanical properties and fracture toughness. Output
routine enables to print and save results as report or
plots and text file appropriate for further analysis.
ACKNOWLEDGMENTS
The authors acknowledge support of European
Commission within the European project Fitness-forPurpose Network for European Industry-FITNET (5th
framework programme)!
REFERENCES
[1] Fatigue Crack Growth Program “NASGO” version
3.0 – reference manual JSC-22267B, NASA
Lyndon B. Johnson Space Center, Houston, 2000.
[2] CRACKWISE® 4 – Automation of Fracture and
Fatigue Assessment Procedures (BS7910) for
Engineering Critical Assessment, TWI Ltd., Granta
Park, Great Abington, Cambridge, 2005.
[3] SINTAP:
Structural
Integrity
Assessment
Procedure – Final reversion, EU-Project BE-1462,
Brite Euram Programme, 1999.
[4] FITNET – European Fitness-for-Service Network,
G1RT-CT-2001-05071, http://www.eurofitnet.org.
[5] Schwalbe, K.-H., Zerbst, U., Kim, Y.-J., Brocks,
W., Cornec, A., Heerens, H. and Amstutz, H.:
EFAM ETM-97 The ETM method for assessing the
significance of crack-like defects in engineering
44 ▪ VOL. 36, No 1, 2008
structures comprising the versins ETM 97/1 and
ETM97/2, GKSS Research Centre Geesthacht,
GKSS 98/E/6, Geesthacht, 1998.
[6] Schwalbe, K.H., Kim, Y.-J., Hao, S., Cornec, A.
and Koçak, M.: EFAM ETM-MM 96: The ETM
Method for Assessing the significance of crack-like
defects in joints with mechanical heterogeneity
(strength mismatch), GKSS Research Centre
Geesthacht, GKSS 97/E/9, Geesthacht, 1997.
[7] R6 Assessment of the Integrity of Structures
Containing Defects, Revision 4, British Energy,
2000.
[8] BS7910:2005 Guide to methods for assessing the
acceptability of flaws in metallic structures, 2005.
FITNET – СОФТВЕРСКИ МОДУЛ ЗА
ПРОЦЕНУ ПОГОДНОСТИ-ЗА-УПОТРЕБУ
Nenad Gubeljak, Mustafa Koçak, Michell Huther,
Tomaz Valh
Овај рад даје концепт FITNET софтвера за модуле
лома. Уобичајено када се изводи оцена структурног
интегритета
компоненти
са
реалном
или
симулираном прслином неопходно је мењати
особине материјала и геометрију прслине. Такав
приступ може обезбедити статистички поуздане
(веродостојне) резултате за максимум оптерећења
капацитета наспрам величине прслине.
FME Transactions